def _broadcast_ragged_targets_for_overlap(target_start, target_limit,
source_splits):
"""Repeats target indices for each source item in the same batch.
Args:
target_start: `<int>[batch_size, (target_size)]`
target_limit: `<int>[batch_size, (target_size)]`
source_splits: `<int64>[batch_size, (source_size+1)]`
Returns:
`<int>[batch_size, (source_size), (target_size)]`.
A tuple of ragged tensors `(tiled_target_start, tiled_target_limit)` where:
* `tiled_target_start[b, s, t] = target_start[b, t]`
* `tiled_target_limit[b, s, t] = target_limit[b, t]`
"""
source_batch_ids = segment_id_ops.row_splits_to_segment_ids(source_splits)
target_start = ragged_tensor.RaggedTensor.from_value_rowids(
ragged_gather_ops.gather(target_start, source_batch_ids),
source_batch_ids)
target_limit = ragged_tensor.RaggedTensor.from_value_rowids(
ragged_gather_ops.gather(target_limit, source_batch_ids),
source_batch_ids)
return (target_start, target_limit)
def _broadcast_ragged_sources_for_overlap(source_start, source_limit,
target_splits):
"""Repeats source indices for each target item in the same batch.
Args:
source_start: `<int>[batch_size, (source_size)]`
source_limit: `<int>[batch_size, (source_size)]`
target_splits: `<int64>[batch_size, (target_size+1)]`
Returns:
`<int>[batch_size, (source_size), (target_size)]`.
A tuple of tensors `(tiled_source_start, tiled_source_limit)` where:
* `tiled_target_start[b, s, t] = source_start[b, s]`
* `tiled_target_limit[b, s, t] = source_limit[b, s]`
"""
source_splits = source_start.row_splits
target_rowlens = target_splits[1:] - target_splits[:-1]
source_batch_ids = segment_id_ops.row_splits_to_segment_ids(source_splits)
# <int64>[sum(source_size[b] for b in range(batch_size))]
# source_repeats[i] is the number of target spans in the batch that contains
# source span i. We need to add a new ragged dimension that repeats each
# source span this number of times.
source_repeats = ragged_gather_ops.gather(target_rowlens, source_batch_ids)
# <int64>[sum(source_size[b] for b in range(batch_size)) + 1]
# The row_splits tensor for the inner ragged dimension of the result tensors.
inner_splits = array_ops.concat([[0], math_ops.cumsum(source_repeats)],
axis=0)
# <int64>[sum(source_size[b] * target_size[b] for b in range(batch_size))]
# Indices for gathering source indices.
source_indices = segment_id_ops.row_splits_to_segment_ids(inner_splits)
source_start = ragged_tensor.RaggedTensor.from_nested_row_splits(
array_ops.gather(source_start.values, source_indices),
[source_splits, inner_splits])
source_limit = ragged_tensor.RaggedTensor.from_nested_row_splits(
array_ops.gather(source_limit.values, source_indices),
[source_splits, inner_splits])
return source_start, source_limit
def value_rowids(self):
"""Returns the row indices for this row partition.
`value_rowids` specifies the row index fo reach value. In particular,
`value_rowids[i]` is the row index for `values[i]`.
Returns:
A 1-D integer `Tensor` with shape `[self.nvals()]`.
The returned tensor is nonnegative, and is sorted in ascending order.
"""
if self._value_rowids is not None:
return self._value_rowids
return segment_id_ops.row_splits_to_segment_ids(self._row_splits)
def value_rowids(self, name=None):
"""Returns the row indices for this row partition.
`value_rowids` specifies the row index fo reach value. In particular,
`value_rowids[i]` is the row index for `values[i]`.
Args:
name: A name prefix for the returned tensor (optional).
Returns:
A 1-D integer `Tensor` with shape `[self.nvals()]`.
The returned tensor is nonnegative, and is sorted in ascending order.
"""
if self._cached_value_rowids is not None:
return self._cached_value_rowids
with ops.name_scope(name, "RaggedValueRowIds", [self]):
return segment_id_ops.row_splits_to_segment_ids(self.row_splits)
def value_rowids(self, name=None):
"""Returns the row indices for this row partition.
Returns a vector with a number of entries equal to nvals, where
the ith value in the tensor indicates the row of the ith value.
Args:
name: A name prefix for the returned tensor (optional).
Returns:
A 1-D integer `Tensor` with shape `self.values.shape[:1]`.
The returned tensor is nonnegative, and is sorted in ascending order.
"""
if self._cached_value_rowids is not None:
return self._cached_value_rowids
with ops.name_scope(name, "RaggedValueRowIds", [self]):
return segment_id_ops.row_splits_to_segment_ids(self.row_splits)
def boolean_mask(data, mask, name=None):
"""Applies a boolean mask to `data` without flattening the mask dimensions.
Returns a potentially ragged tensor that is formed by retaining the elements
in `data` where the corresponding value in `mask` is `True`.
* `output[a1...aA, i, b1...bB] = data[a1...aA, j, b1...bB]`
Where `j` is the `i`th `True` entry of `mask[a1...aA]`.
Note that `output` preserves the mask dimensions `a1...aA`; this differs
from `tf.boolean_mask`, which flattens those dimensions.
Args:
data: A potentially ragged tensor.
mask: A potentially ragged boolean tensor. `mask`'s shape must be a prefix
of `data`'s shape. `rank(mask)` must be known statically.
name: A name prefix for the returned tensor (optional).
Returns:
A potentially ragged tensor that is formed by retaining the elements in
`data` where the corresponding value in `mask` is `True`.
* `rank(output) = rank(data)`.
* `output.ragged_rank = max(data.ragged_rank, rank(mask) - 1)`.
Raises:
ValueError: if `rank(mask)` is not known statically; or if `mask.shape` is
not a prefix of `data.shape`.
#### Examples:
>>> # Aliases for True & False so data and mask line up.
>>> T, F = (True, False)
>>> tf.ragged.boolean_mask( # Mask a 2D Tensor.
... data=[[1, 2, 3], [4, 5, 6], [7, 8, 9]],
... mask=[[T, F, T], [F, F, F], [T, F, F]]).to_list()
[[1, 3], [], [7]]
>>> tf.ragged.boolean_mask( # Mask a 2D RaggedTensor.
... tf.ragged.constant([[1, 2, 3], [4], [5, 6]]),
... tf.ragged.constant([[F, F, T], [F], [T, T]])).to_list()
[[3], [], [5, 6]]
>>> tf.ragged.boolean_mask( # Mask rows of a 2D RaggedTensor.
... tf.ragged.constant([[1, 2, 3], [4], [5, 6]]),
... tf.ragged.constant([True, False, True])).to_list()
[[1, 2, 3], [5, 6]]
"""
with ops.name_scope(name, 'RaggedMask', [data, mask]):
# Convert inputs to tensors.
data = ragged_tensor.convert_to_tensor_or_ragged_tensor(data, name='data')
mask = ragged_tensor.convert_to_tensor_or_ragged_tensor(
mask, dtypes.bool, name='mask')
row_splits_dtype, (data, mask) = ragged_tensor.match_row_splits_dtypes(
data, mask, return_dtype=True)
# Get static rank of mask.
if mask.shape.ndims is None:
raise ValueError('mask.shape.ndims must be known statically.')
elif mask.shape.ndims == 0:
raise ValueError('mask cannot be scalar.')
# If mask is ragged, then recurse with a non-ragged mask.
if ragged_tensor.is_ragged(mask):
if not ragged_tensor.is_ragged(data):
data = ragged_tensor.RaggedTensor.from_tensor(
data,
ragged_rank=mask.ragged_rank,
row_splits_dtype=mask.row_splits.dtype)
# Check that mask.nested_row_splits is a prefix of
# data.nested_row_splits.
splits_list = [
mask.nested_row_splits, data.nested_row_splits[:mask.ragged_rank]
]
with ops.control_dependencies(
ragged_util.assert_splits_match(splits_list)):
# Strip off ragged `splits` until `mask` is non-ragged. Keep the splits
# that we strip off in `splits`, so we can add them back on after
# we recursively mask the non-ragged data.
splits = []
while ragged_tensor.is_ragged(mask):
if mask.shape.ndims > 2:
splits.append(mask.row_splits)
else:
# Count the number of True mask values in each row to find the
# lengths of the filtered rows; then convert to splits.
int_mask = ragged_functional_ops.map_flat_values(
math_ops.cast, mask, dtype=row_splits_dtype)
masked_row_lengths = ragged_math_ops.reduce_sum(int_mask, axis=1)
splits.append(ragged_util.lengths_to_splits(masked_row_lengths))
mask = mask.values
data = data.values
# Recursively apply the nested non-ragged mask to the nested data.
masked_values = boolean_mask(data, mask)
# Add the ragged `splits` back to the result.
masked_values = ragged_tensor.RaggedTensor.from_nested_row_splits(
masked_values, splits, validate=False)
return masked_values
# If mask is non-ragged and has rank 1, and data is ragged, then build a
# ragged tensor with the indicated rows.
elif ragged_tensor.is_ragged(data) and mask.shape.ndims == 1:
# Get the masked splits: first get the length of each row, then filter
# out the rows that we are deleting, and convert that filtered set of
# masks back to a splits tensor.
lengths = data.row_lengths()
masked_lengths = array_ops.boolean_mask(lengths, mask)
masked_splits = ragged_util.lengths_to_splits(masked_lengths)
# Get the masked values: first get row ids corresponding to each
# value, then use tf.gather to build a boolean mask that's false for
# values that come from rows that we are deleting, and use that mask to
# construct the masked values tensor.
segment_ids = segment_id_ops.row_splits_to_segment_ids(data.row_splits)
segment_mask = array_ops.gather(mask, segment_ids)
masked_values = boolean_mask(data.values, segment_mask)
return ragged_tensor.RaggedTensor.from_row_splits(
masked_values, masked_splits, validate=False)
# If mask is non-ragged and has rank>1, then convert it to be ragged,
# with a ragged rank matching data.
if ragged_tensor.is_ragged(data):
mask = ragged_tensor.RaggedTensor.from_tensor(
mask,
ragged_rank=min(data.ragged_rank, mask.shape.ndims - 1),
row_splits_dtype=data.row_splits.dtype)
return boolean_mask(data, mask)
# Otherwise, data and mask are both `Tensor`s.
else:
# Apply `boolean_mask` to get the masked values.
masked_values = array_ops.boolean_mask(data, mask)
if mask.shape.ndims >= 2:
# Add the innermost ragged dimension. For each innermost cell, get the
# number of values it contains. Then flatten that to get a list of
# cell lengths, and convert it to splits. Finally, combine the splits
# and values to get the innermost ragged tensor.
masked_lengths = math_ops.count_nonzero(
mask, axis=-1, dtype=row_splits_dtype)
flattened_masked_lengths = array_ops.reshape(masked_lengths, [-1])
masked_values = ragged_tensor.RaggedTensor.from_row_lengths(
masked_values, flattened_masked_lengths, validate=False)
# Wrap remaining ragged dimensions.
if mask.shape.ndims > 2:
mask_shape = array_ops.shape(mask, out_type=row_splits_dtype)
split_size = math_ops.cumprod(mask_shape) + 1
for dim in range(mask.shape.ndims - 3, -1, -1):
elt_size = mask_shape[dim + 1]
masked_splits = math_ops.range(split_size[dim]) * elt_size
masked_values = ragged_tensor.RaggedTensor.from_row_splits(
masked_values, masked_splits, validate=False)
return masked_values
def _ragged_reduce_aggregate(reduce_op, unsorted_segment_op, rt_input, axis,
name=None):
"""Aggregates across axes of a RaggedTensor using the given `Tensor` ops.
Reduces `rt_input` along the dimensions given in `axis`. The rank of the
tensor is reduced by 1 for each entry in `axis`. If `axis` is not specified,
then all dimensions are reduced, and a scalar value is returned.
This op assumes that `reduce_op` and `unsorted_segment_op` are associative;
if not, then reducing multiple axes will return incorrect results. (In
particular, reducing multiple axes is currently implemented by reducing the
axes one at a time.)
Args:
reduce_op: The tensorflow `op` that should be used to reduce values in
uniform dimensions. Must have the same signature and basic behavior as
`reduce_sum`, `reduce_max`, etc.
unsorted_segment_op: The tensorflow `op` that should be used to combine
values in ragged dimensions. Must have the same signature and basic
behavior as `unsorted_segment_sum`, `unsorted_segment_max`, etc.
rt_input: A `Tensor` or `RaggedTensor` containing the values to be reduced.
axis: The axis or axes to reduce. May be `None` (to reduce all axes), an
`int` (to reduce a single axis), a `list` or `tuple` of `int` (to reduce a
given set of axes), or a `Tensor` with a constant value. Must be in the
range `[0, rt_input.rank)`.
name: A name prefix for the returned tensor (optional).
Returns:
A `RaggedTensor` containing the reduced values. The returned tensor
has the same dtype as `data`, and its shape is given by removing the
dimensions specified in `axis` from `rt_input.shape`. The `ragged_rank`
of the returned tensor is given by substracting any ragged dimensions
specified in `axis` from `rt_input.ragged_rank`.
Raises:
ValueError: If `axis` contains a `Tensor` whose value is not constant.
"""
if not ragged_tensor.is_ragged(rt_input):
return reduce_op(rt_input, axis, name=name)
if isinstance(axis, ops.Tensor):
axis = tensor_util.constant_value(axis)
if axis is None:
raise ValueError('axis must be known at graph construction time.')
# When reducing all axes, just ignore splits & reduce the inner values.
if axis is None:
return reduce_op(rt_input.inner_values, None, name=name)
with ops.name_scope(name, 'RaggedReduce', [rt_input, axis]):
if isinstance(axis, (tuple, list)):
if not axis:
return rt_input
elif len(axis) == 1:
axis = axis[0]
else:
# When reducing multiple axes, just reduce one at a time. This is less
# efficient, and only works for associative ops. (In particular, it
# does not work for reduce_mean.) However, reducing multiple axes at
# once will probably require a nontrivial c++ op.
axis = sorted(axis)
inner_reduced = _ragged_reduce_aggregate(reduce_op, unsorted_segment_op,
rt_input, axis[-1])
return _ragged_reduce_aggregate(reduce_op, unsorted_segment_op,
inner_reduced, axis[:-1])
axis = ragged_util.get_positive_axis(axis, rt_input.shape.ndims)
rt_input = ragged_factory_ops.convert_to_tensor_or_ragged_tensor(
rt_input, name='rt_input')
if axis == 0:
# out[i_1, i_2, ..., i_N] = sum_{j} rt_input[j, i_1, i_2, ..., i_N]
row_lengths = rt_input.row_splits[1:] - rt_input.row_splits[:-1]
num_segments = math_ops.maximum(math_ops.reduce_max(row_lengths), 0)
segment_ids = range(row_lengths).values
return _ragged_segment_aggregate(unsorted_segment_op, rt_input.values,
segment_ids, num_segments)
elif axis == 1:
# out[i_0, i_1, i_2, ..., i_N] = sum_{j} rt_input[i_0, j, i_2, ..., i_N]
num_segments = array_ops.shape(rt_input.row_splits)[0] - 1
segment_ids = segment_id_ops.row_splits_to_segment_ids(
rt_input.row_splits)
return _ragged_segment_aggregate(unsorted_segment_op, rt_input.values,
segment_ids, num_segments)
else:
# out[i_0, ..., i_[axis-1], i_axis+1], ..., i_N] =
# sum_{j} rt_input [i_0, ..., i_[axis-1], j, i_axis+1], ..., i_N]
return rt_input.with_values(
_ragged_reduce_aggregate(reduce_op, unsorted_segment_op,
rt_input.values, axis - 1))
def ragged_reduce_aggregate(reduce_op,
unsorted_segment_op,
rt_input,
axis,
keepdims,
separator=None,
name=None):
"""Aggregates across axes of a RaggedTensor using the given `Tensor` ops.
Reduces `rt_input` along the dimensions given in `axis`. The rank of the
tensor is reduced by 1 for each entry in `axis`. If `axis` is not specified,
then all dimensions are reduced, and a scalar value is returned.
This op assumes that `reduce_op` and `unsorted_segment_op` are associative;
if not, then reducing multiple axes will return incorrect results. (In
particular, reducing multiple axes is currently implemented by reducing the
axes one at a time.)
Args:
reduce_op: The tensorflow `op` that should be used to reduce values in
uniform dimensions. Must have the same signature and basic behavior as
`reduce_sum`, `reduce_max`, etc.
unsorted_segment_op: The tensorflow `op` that should be used to combine
values in ragged dimensions. Must have the same signature and basic
behavior as `unsorted_segment_sum`, `unsorted_segment_max`, etc.
rt_input: A `Tensor` or `RaggedTensor` containing the values to be reduced.
axis: The axis or axes to reduce. May be `None` (to reduce all axes), an
`int` (to reduce a single axis), a `list` or `tuple` of `int` (to reduce a
given set of axes), or a `Tensor` with a constant value. Must be in the
range `[0, rt_input.rank)`.
keepdims: If true, retains reduced dimensions with length 1.
separator: An optional string. Defaults to None. The separator to use when
joining. The separator must not be set for non-string data types. (i.e. if
separator is not None then it uses string ops)
name: A name prefix for the returned tensor (optional).
Returns:
A `RaggedTensor` containing the reduced values. The returned tensor
has the same dtype as `data`, and its shape is given by removing the
dimensions specified in `axis` from `rt_input.shape`. The `ragged_rank`
of the returned tensor is given by substracting any ragged dimensions
specified in `axis` from `rt_input.ragged_rank`.
Raises:
ValueError: If `axis` contains a `Tensor` whose value is not constant.
"""
if not ragged_tensor.is_ragged(rt_input):
if separator is None:
return reduce_op(rt_input, axis, keepdims=keepdims, name=name)
else:
# When separator is not None, We infer that dtype is string and
# reduce_join will be called.
return reduce_op(
rt_input, axis, keepdims=keepdims, name=name, separator=separator)
if isinstance(axis, ops.Tensor):
axis = tensor_util.constant_value(axis)
if axis is None:
raise ValueError('axis must be known at graph construction time.')
if isinstance(axis, np.ndarray):
axis = axis.tolist()
# When reducing all axes, just ignore splits & reduce the inner values.
if axis is None:
result = reduce_op(rt_input.flat_values, None, keepdims=keepdims, name=name)
if keepdims:
# Expand the result to the input number of dimensions.
for _ in rt_input.shape[1:]:
result = array_ops.expand_dims(result, axis=0)
return result
with ops.name_scope(name, 'RaggedReduce', [rt_input, axis]):
if isinstance(axis, (tuple, list)):
if not axis:
return rt_input
elif len(axis) == 1:
axis = axis[0]
else:
# When reducing multiple axes, as we reduce one at a time (see below),
# the negative axis has to be converted to positive at the first run
# as the sort with negative axis will have different orders.
# See GitHub issue 27497.
axis = [
array_ops.get_positive_axis(a, rt_input.shape.ndims, 'axis[%s]' % i,
'rank(input_tensor)')
for i, a in enumerate(axis)
]
# When reducing multiple axes, just reduce one at a time. This is less
# efficient, and only works for associative ops. (In particular, it
# does not work for reduce_mean.) However, reducing multiple axes at
# once will probably require a nontrivial c++ op.
axis = sorted(axis)
inner_reduced = ragged_reduce_aggregate(reduce_op, unsorted_segment_op,
rt_input, axis[-1], keepdims,
separator)
return ragged_reduce_aggregate(reduce_op, unsorted_segment_op,
inner_reduced, axis[:-1], keepdims,
separator)
rt_input = ragged_tensor.convert_to_tensor_or_ragged_tensor(
rt_input, name='rt_input')
axis = array_ops.get_positive_axis(
axis, rt_input.shape.ndims, ndims_name='rank(input_tensor)')
if axis == 0:
# out[i_1, i_2, ..., i_N] = sum_{j} rt_input[j, i_1, i_2, ..., i_N]
row_lengths = rt_input.row_splits[1:] - rt_input.row_splits[:-1]
num_segments = math_ops.maximum(math_ops.reduce_max(row_lengths), 0)
segment_ids = range(row_lengths).values
result = _ragged_segment_aggregate(unsorted_segment_op, rt_input.values,
segment_ids, num_segments, separator)
if keepdims:
result = array_ops.expand_dims(result, axis=0)
return result
elif axis == 1:
# out[i_0, i_1, i_2, ..., i_N] = sum_{j} rt_input[i_0, j, i_2, ..., i_N]
num_segments = array_ops.shape(rt_input.row_splits)[0] - 1
segment_ids = segment_id_ops.row_splits_to_segment_ids(
rt_input.row_splits)
result = _ragged_segment_aggregate(unsorted_segment_op, rt_input.values,
segment_ids, num_segments, separator)
if keepdims:
result = array_ops.expand_dims(result, axis=1)
return result
else:
# out[i_0, ..., i_[axis-1], i_axis+1], ..., i_N] =
# sum_{j} rt_input [i_0, ..., i_[axis-1], j, i_axis+1], ..., i_N]
return rt_input.with_values(
ragged_reduce_aggregate(reduce_op, unsorted_segment_op,
rt_input.values, axis - 1, keepdims,
separator))
def testEmptySplits(self):
# Note: the splits for an empty ragged tensor contains a single zero.
segment_ids = segment_id_ops.row_splits_to_segment_ids([0])
self.assertAllEqual(segment_ids, [])
def testDocStringExample(self):
splits = [0, 3, 3, 5, 6, 9]
expected = [0, 0, 0, 2, 2, 3, 4, 4, 4]
segment_ids = segment_id_ops.row_splits_to_segment_ids(splits)
self.assertAllEqual(segment_ids, expected)
def testEmptySplits(self): # Note: the splits for an empty ragged tensor contains a single zero. segment_ids = segment_id_ops.row_splits_to_segment_ids([0]) self.assertAllEqual(segment_ids, [])
def testDocStringExample(self): splits = [0, 3, 3, 5, 6, 9] expected = [0, 0, 0, 2, 2, 3, 4, 4, 4] segment_ids = segment_id_ops.row_splits_to_segment_ids(splits) self.assertAllEqual(segment_ids, expected)
def _ragged_reduce_aggregate(reduce_op,
unsorted_segment_op,
rt_input,
axis,
keepdims,
name=None):
"""Aggregates across axes of a RaggedTensor using the given `Tensor` ops.
Reduces `rt_input` along the dimensions given in `axis`. The rank of the
tensor is reduced by 1 for each entry in `axis`. If `axis` is not specified,
then all dimensions are reduced, and a scalar value is returned.
This op assumes that `reduce_op` and `unsorted_segment_op` are associative;
if not, then reducing multiple axes will return incorrect results. (In
particular, reducing multiple axes is currently implemented by reducing the
axes one at a time.)
Args:
reduce_op: The tensorflow `op` that should be used to reduce values in
uniform dimensions. Must have the same signature and basic behavior as
`reduce_sum`, `reduce_max`, etc.
unsorted_segment_op: The tensorflow `op` that should be used to combine
values in ragged dimensions. Must have the same signature and basic
behavior as `unsorted_segment_sum`, `unsorted_segment_max`, etc.
rt_input: A `Tensor` or `RaggedTensor` containing the values to be reduced.
axis: The axis or axes to reduce. May be `None` (to reduce all axes), an
`int` (to reduce a single axis), a `list` or `tuple` of `int` (to reduce a
given set of axes), or a `Tensor` with a constant value. Must be in the
range `[0, rt_input.rank)`.
keepdims: If true, retains reduced dimensions with length 1.
name: A name prefix for the returned tensor (optional).
Returns:
A `RaggedTensor` containing the reduced values. The returned tensor
has the same dtype as `data`, and its shape is given by removing the
dimensions specified in `axis` from `rt_input.shape`. The `ragged_rank`
of the returned tensor is given by substracting any ragged dimensions
specified in `axis` from `rt_input.ragged_rank`.
Raises:
ValueError: If `axis` contains a `Tensor` whose value is not constant.
"""
if not ragged_tensor.is_ragged(rt_input):
return reduce_op(rt_input, axis, name=name)
if keepdims:
raise ValueError('keepdims=True is not supported for RaggedTensors.')
if isinstance(axis, ops.Tensor):
axis = tensor_util.constant_value(axis)
if axis is None:
raise ValueError('axis must be known at graph construction time.')
# When reducing all axes, just ignore splits & reduce the inner values.
if axis is None:
return reduce_op(rt_input.inner_values, None, name=name)
with ops.name_scope(name, 'RaggedReduce', [rt_input, axis]):
if isinstance(axis, (tuple, list)):
if not axis:
return rt_input
elif len(axis) == 1:
axis = axis[0]
else:
# When reducing multiple axes, just reduce one at a time. This is less
# efficient, and only works for associative ops. (In particular, it
# does not work for reduce_mean.) However, reducing multiple axes at
# once will probably require a nontrivial c++ op.
axis = sorted(axis)
inner_reduced = _ragged_reduce_aggregate(
reduce_op, unsorted_segment_op, rt_input, axis[-1],
keepdims)
return _ragged_reduce_aggregate(reduce_op, unsorted_segment_op,
inner_reduced, axis[:-1],
keepdims)
axis = ragged_util.get_positive_axis(axis, rt_input.shape.ndims)
rt_input = ragged_factory_ops.convert_to_tensor_or_ragged_tensor(
rt_input, name='rt_input')
if axis == 0:
# out[i_1, i_2, ..., i_N] = sum_{j} rt_input[j, i_1, i_2, ..., i_N]
row_lengths = rt_input.row_splits[1:] - rt_input.row_splits[:-1]
num_segments = math_ops.maximum(math_ops.reduce_max(row_lengths),
0)
segment_ids = range(row_lengths).values
return _ragged_segment_aggregate(unsorted_segment_op,
rt_input.values, segment_ids,
num_segments)
elif axis == 1:
# out[i_0, i_1, i_2, ..., i_N] = sum_{j} rt_input[i_0, j, i_2, ..., i_N]
num_segments = array_ops.shape(rt_input.row_splits)[0] - 1
segment_ids = segment_id_ops.row_splits_to_segment_ids(
rt_input.row_splits)
return _ragged_segment_aggregate(unsorted_segment_op,
rt_input.values, segment_ids,
num_segments)
else:
# out[i_0, ..., i_[axis-1], i_axis+1], ..., i_N] =
# sum_{j} rt_input [i_0, ..., i_[axis-1], j, i_axis+1], ..., i_N]
return rt_input.with_values(
_ragged_reduce_aggregate(reduce_op, unsorted_segment_op,
rt_input.values, axis - 1, keepdims))
def boolean_mask(data, mask, keepdims=False, name=None):
"""Applies a boolean mask to `data`.
Returns a potentially ragged tensor that is formed by retaining the elements
in `data` where the corresponding value in `mask` is `True`.
If `keepdims` is true then outer dimensions (corresponding to the `mask`
dimensions) are preserved, and:
* `output[a1...aA, i, b1...bB] = data[a1...aA, j, b1...bB]`
Where `j` is the `i`th `True` entry of `mask[a1...aA]`.
If `keepdims` is false, then the outer dimensions are collapsed (similar to
the behavior of `tf.boolean_mask`), and:
* `output[i, b1...bB] = data[a1...aA, b1...bB]`
Where `(a1...aA)` is the `i`th `True` entry of `mask`
(in row-major order).
Args:
data: A potentially ragged tensor.
mask: A potentially ragged boolean tensor. `mask`'s shape must be a prefix
of `data`'s shape. `rank(mask)` must be known statically.
keepdims: Whether to preserve the outer dimensions (`keepdims=True`) or
flatten them (`keepdims=False`).
name: A name prefix for the returned tensor (optional).
Returns:
A potentially ragged tensor that is formed by retaining the elements in
`data` where the corresponding value in `mask` is `True`.
If `keepdims` is false:
* `rank(output) = rank(data) - rank(mask) + 1`.
* `output.ragged_rank = max(data.ragged_rank - rank(mask) + 1, 0)`.
If `keepdims` is true:
* `rank(output) = rank(data)`.
* `output.ragged_rank = max(data.ragged_rank, rank(mask) - 1)`.
Raises:
ValueError: if `rank(mask)` is not known statically; or if `mask.shape` is
not a prefix of `data.shape`.
#### Examples:
```python
>>> # Aliases for True & False so data and mask line up.
>>> T, F = (True, False)
>>> tf.ragged.boolean_mask( # Mask a 2D Tensor. Flatten outer dims.
... data=[[1, 2, 3], [4, 5, 6], [7, 8, 9]],
... mask=[[T, F, T], [F, F, F], [T, F, F]],
... keepdims=False).tolist()
[1, 3, 7]
>>> tf.ragged.boolean_mask( # Mask a 2D Tensor. Preserve outer dims.
... data=[[1, 2, 3], [4, 5, 6], [7, 8, 9]],
... mask=[[T, F, T], [F, F, F], [T, F, F]],
... keepdims=True).tolist()
[[1, 3], [], [7]]
>>> tf.ragged.boolean_mask( # Mask a 2D RaggedTensor. Flatten outer dims.
... tf.ragged.constant([[1, 2, 3], [4], [5, 6]]),
... tf.ragged.constant([[F, F, T], [F], [T, T]]),
... keepdims=False).tolist()
[3, 5, 6]
>>> tf.ragged.boolean_mask( # Mask a 2D RaggedTensor. Preserve outer dims.
... tf.ragged.constant([[1, 2, 3], [4], [5, 6]]),
... tf.ragged.constant([[F, F, T], [F], [T, T]]),
... keepdims=True).tolist()
[[3], [], [5, 6]]
>>> tf.ragged.boolean_mask( # Mask rows of a 2D RaggedTensor.
... tf.ragged.constant([[1, 2, 3], [4], [5, 6]]),
... tf.ragged.constant([True, False, True]),
... keepdims=True).tolist()
[[1, 2, 3], [5, 6]]
```
"""
with ops.name_scope(name, 'RaggedMask', [data, mask]):
# Convert inputs to tensors.
data = ragged_tensor.convert_to_tensor_or_ragged_tensor(data, name='data')
mask = ragged_tensor.convert_to_tensor_or_ragged_tensor(
mask, dtypes.bool, name='mask')
row_splits_dtype, (data, mask) = ragged_tensor.match_row_splits_dtypes(
data, mask, return_dtype=True)
# Get static rank of mask.
if mask.shape.ndims is None:
raise ValueError('mask.shape.ndims must be known statically.')
elif mask.shape.ndims == 0:
raise ValueError('mask cannot be scalar.')
# If mask is ragged, then recurse with a non-ragged mask.
if ragged_tensor.is_ragged(mask):
if not ragged_tensor.is_ragged(data):
data = ragged_tensor.RaggedTensor.from_tensor(
data, ragged_rank=mask.ragged_rank,
row_splits_dtype=mask.row_splits.dtype)
# Check that mask.nested_row_splits is a prefix of
# data.nested_row_splits.
splits_list = [
mask.nested_row_splits, data.nested_row_splits[:mask.ragged_rank]
]
with ops.control_dependencies(
ragged_util.assert_splits_match(splits_list)):
# Strip off ragged `splits` until `mask` is non-ragged. Keep the splits
# that we strip off in `splits`, so we can add them back on after
# we recursively mask the non-ragged data.
splits = []
while ragged_tensor.is_ragged(mask):
if mask.shape.ndims > 2:
splits.append(mask.row_splits)
else:
# Count the number of True mask values in each row to find the
# lengths of the filtered rows; then convert to splits.
int_mask = ragged_functional_ops.map_flat_values(
math_ops.cast, mask, dtype=row_splits_dtype)
masked_row_lengths = ragged_math_ops.reduce_sum(int_mask, axis=1)
splits.append(ragged_util.lengths_to_splits(masked_row_lengths))
mask = mask.values
data = data.values
# Recursively apply the nested non-ragged mask to the nested data.
masked_values = boolean_mask(data, mask, keepdims)
# Add the ragged `splits` back to the result.
if keepdims:
masked_values = ragged_tensor.RaggedTensor.from_nested_row_splits(
masked_values, splits, validate=False)
return masked_values
# If mask is non-ragged and has rank 1, and data is ragged, then build a
# ragged tensor with the indicated rows.
elif ragged_tensor.is_ragged(data) and mask.shape.ndims == 1:
# Get the masked splits: first get the length of each row, then filter
# out the rows that we are deleting, and convert that filtered set of
# masks back to a splits tensor.
lengths = data.row_lengths()
masked_lengths = array_ops.boolean_mask(lengths, mask)
masked_splits = ragged_util.lengths_to_splits(masked_lengths)
# Get the masked values: first get row ids corresponding to each
# value, then use tf.gather to build a boolean mask that's false for
# values that come from rows that we are deleting, and use that mask to
# construct the masked values tensor.
segment_ids = segment_id_ops.row_splits_to_segment_ids(data.row_splits)
segment_mask = array_ops.gather(mask, segment_ids)
masked_values = boolean_mask(data.values, segment_mask, keepdims=False)
return ragged_tensor.RaggedTensor.from_row_splits(masked_values,
masked_splits,
validate=False)
# If mask is non-ragged and has rank>1, then convert it to be ragged,
# with a ragged rank matching data.
if ragged_tensor.is_ragged(data):
mask = ragged_tensor.RaggedTensor.from_tensor(
mask, ragged_rank=min(data.ragged_rank, mask.shape.ndims - 1),
row_splits_dtype=data.row_splits.dtype)
return boolean_mask(data, mask, keepdims)
# Otherwise, data and mask are both `Tensor`s.
else:
# Apply `boolean_mask` to get the masked values.
masked_values = array_ops.boolean_mask(data, mask)
if mask.shape.ndims >= 2 and keepdims:
# Add the innermost ragged dimension. For each innermost cell, get the
# number of values it contains. Then flatten that to get a list of
# cell lengths, and convert it to splits. Finally, combine the splits
# and values to get the innermost ragged tensor.
masked_lengths = math_ops.count_nonzero(mask, axis=-1,
dtype=row_splits_dtype)
flattened_masked_lengths = array_ops.reshape(masked_lengths, [-1])
masked_values = ragged_tensor.RaggedTensor.from_row_lengths(
masked_values, flattened_masked_lengths, validate=False)
# Wrap remaining ragged dimensions.
if mask.shape.ndims > 2 and keepdims:
mask_shape = array_ops.shape(mask, out_type=row_splits_dtype)
split_size = math_ops.cumprod(mask_shape) + 1
for dim in range(mask.shape.ndims - 3, -1, -1):
elt_size = mask_shape[dim + 1]
masked_splits = math_ops.range(split_size[dim]) * elt_size
masked_values = ragged_tensor.RaggedTensor.from_row_splits(
masked_values, masked_splits, validate=False)
return masked_values
